special section the international mobility of inventors … · special section the international...

SPECIAL SECTION THE INTERNATIONAL MOBILITY OF INVENTORS

21

Introduction

The relationship between migration and innovation has

become a major focus of research by academics and pol-

icymakers alike. The key factor driving this development

is the observation that high-skilled migrants decisively

contribute to innovation outcomes, to the international

diffusion of knowledge and, ultimately, to the economic

growth of nations.

In some of the largest migrant-receiving countries (e.g.,

the United States of America (US)), immigrants are over-

represented among the most skilled workers. While immi-

grants account for about 12% of the entire US labor force,

they account for 25% of US scientists and engineers,

50% of US PhDs, 60% of post-doctoral students, and

26% of US-based Nobel Laureates (Black and Stephan,

2008; Kerr, 2009). Some anecdotal evidence suggests

that this overrepresentation of immigrants among high-

skilled workers is not unique to the US, but extends to

other countries that receive large numbers of migrants

(Fink et al, 2013). Thus, an increasing, albeit still limited,

number of studies have linked high-skilled immigration to

knowledge creation (see Breschi et al, 2013; Kerr, 2013,

for recent surveys). Given this situation, many countries

are currently debating and reforming their immigration

policies. A key question governments and policy makers

face is how to attract skilled workers who can relieve

domestic skills shortages and foster innovation.

This special section discusses the opportunities for using

IP data and patent applications, in particular, for migration

related research. It does so by describing the main pat-

terns and trends in inventor international migration – data

which were elicited from information contained in Patent

Cooperation Treaty (PCT) applications. The next section

briefly describes the source of the data, while the follow-

ing sections more extensively analyze aggregated figures

on the phenomenon of inventor migration and explore

the possibilities of using these data for future research.

What can patent data tell us about skilled migration?

The literature on migration and innovation is limited, mainly

due to the relative lack of data that have characterized this

research field. In the last 15 years, census-based migra-

tion datasets have been the data source most commonly

used to conduct research on migration issues as well as

to study the migration-innovation nexus. These datasets

comprise information on migrants by destination country

based on population censuses. Notwithstanding their

value for economic research, census-based datasets

have certain limitations. For example, the data are only

released every 10 years. Moreover, the majority of existing

datasets provide a skills breakdown according to three

schooling levels: primary, secondary and tertiary, which

only offers a rough differentiation of skills.

SPECIAL SECTIONTHE INTERNATIONAL MOBILITY OF INVENTORS


22

More recently, information retrieved from patent docu-

ments has also been used for the purpose of undertaking

innovation-migration research. Broadly speaking, patent

applications contain relevant information on the inventors

and owners of the patent, including the inventors’/owners’

names and addresses, technologies (IPC classifications)

and backward citations. Thus, patent data are an unri-

valled indicator for studying a number of innovation-re-

lated phenomena, such as the mobility of inventors, their

social networks and the patterns of knowledge diffusion.

The potential benefits of using inventor migration data

as captured in patent applications - which this section

elucidates - are manifold. First, data are related to one

specific class of high-skilled workers that are bound to be

more homogenous than the group of tertiary-educated

workers as a whole. In addition, inventors arguably have

special economic importance, as they create knowledge

that is at the genesis of technological and industrial trans-

formation. The use of patent-inventor data for migration

analysis implies the direct measurement of migrants,

contribution to innovation in their destination countries’.

Finally, patent data (and therefore inventor-related infor-

mation) are collected on a yearly basis, and such data are

available for a large number of “sending” and “receiving

countries” at a relatively low cost.

Recently, scholars have undertaken studies of migrant

inventors using information from patent applications

(Breschi et al, 2013; Kerr, 2009). In particular, they have

sought to identify the likely cultural origin of inventor

names disclosed in patent data, which provides import-

ant insights. However, the cultural origin of inventor names

may not always indicate recent migratory background

– for example, Turkish immigrants in Germany.

PCT applications contain information on the nationality

of inventors as well as information on their country of

residence at (for a detailed description of the data source,

see Miguélez and Fink, 2013). This information is available

due to one of the requirements under the PCT specifying

that only nationals or residents of a PCT contracting

state can file PCT applications. To verify that applicants

meet at least one of the two eligibility criteria, the PCT

application form requires applicants to provide details

of both their nationality and their residency. Moreover, it

transpires that, until 2012, US patent application proce-

dures have required all inventors in PCT applications to

be listed as applicants. Thus, if a given PCT application

included the US as a country in which the applicant was

considering pursuing a patent – a so-called designated

state in the patent application – all inventors were listed

as applicants, whereby ensuring that information on their

residence and nationality were available. The majority of

PCT applicants seek protection in the US, reflecting the

popularity of this country as the world’s largest market.

As a result, these data offer a valuable resource to bet-

ter understanding high-skilled migration flows and their

implications for innovation.1

The PCT database comprises more than 6 million names

of the inventors detailed in PCT applications. These

names include some homonyms which may (or may

not) refer to the same inventor. The database does not,

however, provide a single identifier for each inventor,

which makes it difficult to consolidate inventor names.

For example, when two applications contain identical

inventor names, it is difficult to distinguish whether they

are filed by the same inventor or by two different inventors.

1 Unfortunately, the US enacted changes to its patent

laws under the Leahy-Smith America Invents Act

(AIA), which effectively removed the requirement

that inventors also be named as applicants.

Starting on September 16, 2012, PCT applicants

(automatically) designating the US became free to

list inventors and are no longer obliged to indicate

their nationality and residence. As a result, many

applicants do not provide such information any longer.


23

The economic literature has disambiguated individual

inventors through their names and surnames as well as

through other information contained in patent documents.

This section does not attempt to disambiguate inventor

names, and it treats each combination of the inventor

name with an application number as if it were a different

inventor. Although this approach is far from perfect, it

enables meaningful analysis on an aggregate level.

Overall, the share of PCT data with information on na-

tionality and residency was very high, i.e., approximately

80% for the 1978-2012 period. However, this coverage

was unevenly distributed over time – approximately 60-

70% during the 1990s and 70-95% during the 2000s.

Coverage was also unevenly distributed across coun-

tries: US (66%), Canada (81%), the Netherlands (74%),

Germany (95%), the United Kingdom (UK, 92%), France

(94%), Switzerland (93%), China (92%) and India (90%),

among others.

Using the inventor’s nationality information outlined above,

the following subsections present several migration-relat-

ed figures. These figures clearly show that the pattern of

inventor’s mobility, especially from the perspective of the

receiving countries, resembles other high-skilled migra-

tion figures, and in particular, what is known about the

migration of scientists and engineers based on anecdotal

evidence, surveys and media reports.

Where do migrant inventors emigrate to/come from?

Analysis of all records containing complete information

has shown that approximately 5 million, i.e., 9-10% of

inventors had a migration background – i.e., their place

of residence was different from their nationality. This

share has increased over time – it was 7.8% during the

1996-2000 period and 10.1% during the 2006-10 period.

Immigrant inventors were overwhelmingly concentrated

in high-income countries, both during the 1996-2000

and 2006-10 periods (see Table 1). North America ac-

counted for the highest concentration of immigrant

inventors in high-income economies. During the period

2006-10, 59.1% of immigrant inventors were residing

in North America – which is a share that is larger than

that recorded during the late 1990s. A total of 31.4% of

immigrant inventors lived in Europe over the 2006-10

period, which is lower than its 1996-2000 share. Asia

lagged far behind, accounting for 7.5% of all immigrant

inventors during the 2006-10 period.

Table 1: Shares of immigrant and emigrant inventors by income group and region: 1996-2000 and 2006-10

Income group / Region

Immigrant inventors (%)1996-2000

Immigrant inventors (%)

2006-10

Emigrant inventors (%) 1996-2000

Emigrant inventors (%)

2006-10Income groupHigh-income 98.1 97.2 66.9 57.7

Upper middle-income 1.7 2.4 22.2 26.8

Lower middle-income 0.2 0.3 10.4 14.9

Low-income 0.1 0.1 0.5 0.6

RegionAfrica 0.5 0.1 1.8 1.7

Asia 5.0 7.5 31.8 41.9

Europe 39.3 31.4 52.0 41.9Latin America and the Caribbean 0.7 0.3 2.2 2.7

North America 51.5 59.1 9.7 9.7

Oceania 3.1 1.5 2.5 2.1

Note: Income groups are defined according to the World Bank classification, 2012. Source: WIPO Statistics Database, October 2013


24

Table 1 also presents the data from the perspective of

the sending countries. The first interesting point to note

is that the largest proportion of out-migration of inventors

also occurred in high-income countries. However, the

share of inventor emigrants from these countries was

considerably lower when compared to the share of

inventor immigrants. Indeed, middle-income countries

accounted for more than 40% of emigrant inventors

during the 2006-10 period. Moreover, when the data for

the 1996-2000 and 2006-10 time periods are compared,

it is possible to see that the contribution of middle-income

economies increased considerably – i.e., approximately

nine percentage points – while the corresponding share

for high-income countries decreased by the same order

of magnitude.

Like immigration, emigration was highly concentrated in

two world regions, namely, Asia and Europe. Together

these two regions accounted for more than 83% of

inventor emigrants during the period 2006-10.2

2 It should be noted that from the 1996-2000 period

to 2006-10 period, the share of emigrant inventors

from Asian countries increased considerably i.e., from

31.8% to 41.9%, while the share of European emigrant

inventors decreased by approximately 10 percentage

points between the same time periods (see Table 1).

Table 2 provides immigrant and emigrant data broken

down by country. The majority of immigrant inventors

were concentrated in the US, which accounted for 57.1%

of all inventors during the 2006-10 period. European

countries, such as France, Germany, Switzerland, the

Netherlands and the UK, lagged far behind.

As can be observed, other high-income countries also

accounted for large numbers of emigrant inventors;

indeed, such countries were ranked among the top 20

in terms of having the largest emigrant communities.

However, for the 2006-10 period, China and India topped

the world ranking, followed by Germany and the UK.

When compared with immigration patterns, emigrant

inventors were more evenly distributed across countries.

On the one hand, the US alone received approximately

57% of all immigrant inventors; on the other, six coun-

tries (Canada, China, France, Germany, India and the

UK) hosted 57% of all emigrant inventors. Interestingly,

countries such as Canada, France, Germany and the

UK, despite being critical attractors of talent, saw more

inventors emigrating than immigrating.


25

Table 2: Top 20 countries with the largest inventor immigrant and emigrant communities, 2006-10

Country ImmigrantsShare of

world total (%) Country EmigrantsShare over

world total (%)United States of America 117,244 57.1 China 33,413 16.3

Germany 14,547 7.1 India 24,807 12.1

Switzerland 12,479 6.1 Germany 19,043 9.3

United Kingdom 9,113 4.4 United Kingdom 15,160 7.4

Netherlands 5,565 2.7 Canada 13,056 6.4

France 5,369 2.6 France 11,790 5.7

Singapore 4,334 2.1 United States of America 6,795 3.3

Canada 4,107 2.0 Republic of Korea 6,101 3.0

Japan 4,092 2.0 Italy 6,092 3.0

China 3,289 1.6 Netherlands 5,052 2.5

Sweden 3,204 1.6 Russian Federation 4,404 2.1

Belgium 3,173 1.5 Japan 4,029 2.0

Australia 2,441 1.2 Australia 3,212 1.6

Finland 1,969 1.0 Spain 3,085 1.5

Austria 1,905 0.9 Austria 2,775 1.4

Spain 1,590 0.8 Sweden 2,506 1.2

Denmark 1,520 0.7 Israel 2,252 1.1

Republic of Korea 1,188 0.6 Turkey 2,046 1.0

Italy 1,108 0.5 Belgium 1,932 0.9

Ireland 1,092 0.5 Greece 1,886 0.9

World 205,446 100 World 205,446 100

Source: WIPO Statistics Database, October 2013

Figure 1: Inventor immigration rates for the largest receiving countries, 2006-10

Inventor immigration rate (%): 1996-2000

44.3 28.6 15.8 17.0 16.9 14.9 8.6 9.8 8.2 10.8 5.6 10.8 6.2 3.4 5.4 6.0 5.1 3.9 3.9 4.0 2.5 0.8 1.0 0.6

52.9

40.4

20.7 19.9 18.5 17.414.7 12.9 12.9 12.0 11.1 10.9 10.5 10.1 9.5

6.8 6.2 5.83.0 2.8 1.8 1.7 1.2 1.0In

vent

or im

mig

ratio

n ra

te (%

): 20

06-1

0

Singa

pore

Switz

erlan

d

Irelan

d

Belgi

um

United

State

s of A

merica

New Ze

aland

Netherl

ands

Austria

United

King

dom

Austral

ia

Norway

Canad

a

Denmark

Finlan

d

Swed

enSp

ainFra

nce

German

yIta

lyChin

aIsr

ael

India

Japan

Repu

blic o

f Kore

a

Receiving country



26

The US accounted for not only the largest absolute

number of immigrant inventors, but it also had a high

immigration rate of inventors, which is defined as total

number of immigrant inventors over the total number

of inventors (Figure 1). However, during the 2006-10

period, Singapore (52.9%) had the highest immigration

rate, followed by Switzerland (40.4%), Ireland (20.7%)

and Belgium (19.9%). Figure 1 also shows inventor

immigration rates for the 1996-2000 period. Countries

such as Denmark, Finland, Germany, Norway, Sweden,

Switzerland, the Netherlands, and the UK recorded con-

siderable increases in their immigration rates between

the 1996-2000 and the 2006-10 periods.

Who recruits internationally?

In general, inventor immigration rates differ not only

across countries, but also across different applicants.

For example, Table 3 lists the immigration rates for the

top 10 PCT applicants – based on the residence of

the first-named applicant for the 2006-10 period for a

selection of countries. It shows that the distribution of

immigrant inventors was very uneven across applicants,

even between enterprises of a relatively similar size. In

France, for example, France Telecom’s rate of immigrant

inventors was between four and five times greater than

that of Peugeot-Citroen – an imbalance which cannot

be solely attributed to differences across technology

fields. In another example, Peugeot-Citroen, had an

immigration rate that was more than ten times greater

than that of Renault S.A.S.

One interesting aspect of the data highlighted in Table

3 is the role played by universities and public research

centers in the recruitment of talent from abroad. The

top patenting universities and public research centers

feature some of the highest inventor immigration rates

among the top PCT applicants. This is the case for the

University of California in the US, for example, and also

for Cambridge University, Imperial Innovations (Imperial

College London), and Isis Innovation (Oxford University)

in the UK, among others.


27

Table 3: Inventor immigration rates for top 10 applicants, selected countries, 2006-10

Applicant's nameImmigration

rate (%) Applicant Inventor

United States of AmericaQUALCOMM INCORPORATED 50.8 6,528 19,907

MICROSOFT CORPORATION 57.4 3,020 11,297

3M INNOVATIVE PROPERTIES COMPANY 11 2,577 8,852HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P. 18.6 2,360 6,114

E.I. DUPONT DE NEMOURS AND COMPANY 17 2,118 5,916INTERNATIONAL BUSINESS MACHINES CORPORATION 21.4 2,006 6,854

UNIVERSITY OF CALIFORNIA 28.2 1,754 5,598

MOTOROLA, INC. 23.4 1,573 4,488

PROCTER & GAMBLE COMPANY 10.2 1,540 4,953

BAKER HUGHES INCORPORATED 12.8 1,461 3,552

SwitzerlandNESTEC S.A. 56.4 619 1,781

F. HOFFMANN-LA ROCHE AG 46.6 564 1,385

NOVARTIS AG 62.6 489 1,179

SYNGENTA PARTICIPATIONS AG 66.6 308 972

ACTELION PHARMACEUTICALS LTD 30.2 272 879

ALSTOM TECHNOLOGY LTD 67.6 212 506

ABB RESEARCH LTD 65 201 529SWISS FEDERAL INSTITUTE OF TECHNOLOGY 49.2 186 534

SIKA TECHNOLOGY AG 30.4 179 426

INVENTIO AG 23.6 174 338

SingaporeAGENCY OF SCIENCE, TECHNOLOGY AND RESEARCH 62.2 791 2,690

NATIONAL UNIVERSITY OF SINGAPORE 57.6 213 735

NANYANG TECHNOLOGICAL UNIVERSITY 61.4 148 474

CREATIVE TECHNOLOGY LTD 21.6 88 217

NANYANG POLYTECHNIC 23 74 166

SINGAPORE HEALTH SERVICES PTE LTD 37.4 35 160TEMASEK LIFE SCIENCES LABORATORY LIMITED 70.6 28 78

RAZER (ASIA-PACIFIC) PTE LTD 4.6 27 44

SIEMENS MEDICAL INSTRUMENTS PTE. LTD. 25 27 76

S*BIO PTE LTD 77.6 17 49

ChinaZTE CORPORATION 0.2 7,551 17,803

HUAWEI TECHNOLOGIES CO., LTD. 0.8 7,277 18,858

HUAWEI DEVICE CO., LTD. 0.2 570 1,372TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED 0 419 1,014

ALCATEL SHANGHAI BELL CO., LTD. 0.4 380 1,095CHINA ACADEMY OF TELECOMMUNICATIONS TECHNOLOGY 2 317 1,002

BYD COMPANY LIMITED 0 263 1,015

TSINGHUA UNIVERSITY 0.2 242 1,571

PEKING UNIVERSITY 0.2 215 818DA TANG MOBILE COMMUNICATIONS EQUIPMENT CO., LTD. 0.6 205 688


Applicant's nameImmigration

rate (%) Applicant Inventor

GermanyROBERT BOSCH CORPORATION 2.8 6,480 17,484

SIEMENS AKTIENGESELLSCHAFT 6.4 4,555 11,753

BASF SE 14.4 3,562 15,427

BOSCH-SIEMENS HAUSGERATE GMBH 3.2 1,679 4,575FRAUNHOFER-GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG E.V. 5.4 1,532 5,521

CONTINENTAL AUTOMOTIVE GMBH 8.6 1,337 3,447HENKEL KOMMANDITGESELLSCHAFT AUF AKTIEN 6.4 1,210 4,420

DAIMLER AG 3.8 1,196 3,601

EVONIK DEGUSSA GMBH 5.6 974 4,103

ZF FRIEDRICHSHAFEN AG 2.4 958 2,702

United KingdomUNILEVER PLC 10.4 594 1,536

GLAXO GROUP LIMITED 12.6 409 1,590BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY 20.2 389 861

BAE SYSTEMS PLC 3.2 305 644

IMPERIAL INNOVATIONS LTD. 29.8 246 648

ISIS INNOVATION LIMITED 29.8 242 618

DYSON TECHNOLOGY LIMITED 10.4 237 579

ASTRAZENECA UK LIMITED 8.2 210 640

CAMBRIDGE UNIVERSITY 36.6 205 572

QINETIQ LIMITED 2.2 185 458

FranceCENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE (CNRS) 8 1,892 7,002COMMISSARIAT A L'ENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES 2.6 1,514 4,240

RENAULT S.A.S. 0.2 1,065 2,357

FRANCE TELECOM 11.6 963 2,188

L'OREAL 1.8 849 1,730

PEUGEOT CITROEN AUTOMOBILES SA 2.4 772 1,502

THALES ULTRASONICS SAS 0.4 626 1,473INSTITUT NATIONAL DE LA SANTE ET DE LA RECHERCHE MEDICALE (INSERM) 9.2 517 1,633

ARKEMA 3.4 506 1,279L AIR LIQUIDE SOCIETE ANONYME POUR L'ETUDE ET L'EXPLOITATION DES PROCEDES GEORGES CLAUDE 5 471 1,332

IndiaCOUNCIL OF SCIENTIFIC AND INDUSTRIAL RESEARCH 0 304 1,477

HINDUSTAN UNILEVER LIMITED 1.4 178 602

RANBAXY LABORATORIES LIMITED 1.8 161 793

DR. REDDY'S LABORATORIES LTD. 0.8 134 891

CADILA HEALTHCARE LIMITED 0.8 128 455

LUPIN LIMITED 3.8 117 564

MATRIX LABORATORIES LTD 0 97 535

CIPLA LIMITED 0 87 257

INDIAN INSTITUTE OF TECHNOLOGY 0.6 82 200

WOCKHARDT LIMITED 1 75 323


28

What are inventors’ preferred entrance routes?

Even if the evidence is only anecdotal, it seems reasonable

to argue that universities and public research organiza-

tions act as privileged “points of entry” for high-skilled

workers from abroad. Figure 2 explores this scenario by

depicting inventor immigration rates across countries, bro-

ken down by four types of applicants: university; govern-

ment and research institutions; business, and individuals.

Bearing in mind that the business sector accounts for the

vast majority (over 80%) of PCT applications (WIPO, 2012)

in most of the countries listed in Figure 2, the university

and government sectors accounted for the highest immi-

gration rates. In selected cases, the university/government

immigration rates were considerably higher than the busi-

ness immigration rates – in particular, in Australia, Canada,

Japan, Norway, Sweden, the Republic of Korea, the UK

and the US. Only Belgium, China, Finland, India, Italy, the

Netherlands and Spain did not report higher immigration

rates for inventors working in academic institutions, as

opposed to those working in commercial enterprises.

Figure 2: Immigration rates of inventors by type of applicant: business, university, research/government, and individual, 2006-10

0

20

40

60

Inve

ntor

imm

igra

tion

rate

(%)

Singa

pore

Switz

erlan

d

New Ze

aland

United

State

s of A

merica

Irelan

d

Norway

United

King

dom

Swed

en

Belgi

um

Austral

ia

Austria

Canad

a

Netherl

ands

Denmark

Spain

Franc

e

German

y

Finlan

dIta

lyJap

anChin

aIsr

ael

Repu

blic o

f Kore

aInd

ia

Receiving country

University Research/government Business Individual


Do inventor immigration rates differ across technological fields?

As is apparent from analysis of applicant-level data, immi-

grant inventors’ contribution to patenting differ markedly

across technology fields. For example, inventors may

be associated with one or more International Patent

Classification (IPC) symbols, which in turn are grouped

into 35 technology fields through the concordance table

developed by WIPO.3 It should be noted that when a

PCT application relates to multiple fields of technology,

3 WIPO has developed a concordance table in

order to link IPC symbols to corresponding fields

of technology (see www.wipo.int/ipstats/en).

the inventor is counted twice. Therefore, adding up the

absolute numbers of inventors across the 35 technology

fields results in a larger number of inventors than that

outlined earlier in this report.

The 35 fields can be divided into broader technology

groupings – electrical engineering, instruments, chemistry,

mechanical engineering and others. As shown in Figure 3,

all technology fields have recorded increases in the rates

of immigration during the 1990-2010 period. However,

electrical engineering and chemistry emerge as the most

attractive sectors for foreign inventors. In contrast, the

field of mechanical engineering has remained more or

less stable.


29

Figure 3: Inventor immigration rates over time by field of technology: three-year moving averages

4

6

8

10

12

14

Inve

ntor

imm

igra

tion

rate

(%)

1990 1995 2000 2005 2010

Year

Electrical engineering Instruments ChemistryMechanical engineering Others


Table 4 shows inventor immigration rates by field of

technology for the 1996-2000 and 2006-10 periods. As

can be seen, the differences across technology fields –

in terms of how they relied on foreign inventors – were

noticeable. Thus, for example, during the 2006-10 period,

immigration rates varied from 4.1% (mechanical elements)

to 18.3% (micro-structure and nano-technology). Other

fields also relied heavily on immigrant inventors; such

fields included pharmaceuticals (14.6%), biotechnology

(14.6%), digital communication (15.2%), and basic com-

munication processes (16%). The majority of technology

fields had a higher inventor immigration rate for the

2006-10 period compared to the 1996-2000 period.

Despite a decrease, both analyses of biomaterials and

biotechnology fields showed a high inventor immigration

rate for both periods.

Table 4: Inventor immigration rates by technology field, 2006-10

Field of technology

Immigration rate (%),

1996-2000

Immigration rate (%), 2006-10

Electrical engineeringElectrical machinery, energy 5.2 7.2

Audio-visual technology 6.2 9.5

Telecommunications 7.5 11.9

Digital communication 9.7 15.2

Basic communication processes 9.2 16.0

Computer technology 9.6 13.4

IT methods for management 8.0 10.5

Semiconductors 7.0 12.1

InstrumentsOptics 6.5 7.9Measurement 7.0 9.8

Analysis of biological materials 13.9 13.8

Control apparatus 5.3 7.0

Medical technology 6.9 8.3

ChemistryOrganic fine chemistry 9.3 13.9

Biotechnology 16.5 14.6

Pharmaceuticals 11.3 14.6

Macromolecular chemistry, polymers 7.2 10.2

Food chemistry 7.9 11.2

Basic materials chemistry 7.6 11.4

Materials metallurgy 5.7 7.7

Surface technology, coating 5.9 8.1

Micro-structure and nano-technology 13.0 18.3

Chemical engineering 6.5 9.0

Environmental technology 4.6 7.3

Mechanical engineeringHandling 4.5 5.1

Machine tools 3.6 4.6

Engines, pumps, turbines 4.4 6.1

Textile and paper 5.1 6.8

Other special machines 5.0 6.4

Thermal processes and apparatus 4.3 5.2

Mechanical elements 3.8 4.1

Transport 3.9 4.3

Other fieldsFurniture, games 4.7 5.0

Other consumer goods 5.4 5.3

Civil engineering 4.4 7.7

Note: The IPC-technology concordance table (available at: www.wipo.int/ipstats/en) was used to convert IPC symbols into 35 corresponding fields of technology.



30

Figure 4: Inventor immigration rates for selected technology fields and countries, 2006-10

0

10

20

30

40

Inve

ntor

imm

igra

tion

rate

(%)

Switz

erlan

d

Netherl

ands

United

State

s of A

merica

United

King

dom

Canad

aFra

nce

German

yChin

aInd

iaJap

an

Electrical machinery, energy

0

10

20

30

40

50

Inve

ntor

imm

igra

tion

rate

(%)

Switz

erlan

d

Netherl

ands

United

State

s of A

merica

United

King

dom

Canad

aFra

nce

German

yChin

aInd

iaJap

an

Digital communication

0

10

20

30

40

Inve

ntor

imm

igra

tion

rate

(%)

Switz

erlan

d

Netherl

ands

United

State

s of A

merica

United

King

dom

Canad

aFra

nce

German

yChin

aInd

iaJap

an

Computer technology

0

10

20

30

Inve

ntor

imm

igra

tion

rate

(%)

Switz

erlan

d

Netherl

ands

United

State

s of A

merica

United

King

dom

Canad

aFra

nce

German

yChin

aInd

iaJap

an

Measurement

0

10

20

30

Inve

ntor

imm

igra

tion

rate

(%)

Switz

erlan

d

Netherl

ands

United

State

s of A

merica

United

King

dom

Canad

aFra

nce

German

yChin

aInd

iaJap

an

Medical technology

0

10

20

30

40

50

Inve

ntor

imm

igra

tion

rate

(%)

Switz

erlan

d

Netherl

ands

United

State

s of A

merica

United

King

dom

Canad

aFra

nce

German

yChin

aInd

iaJap

an

Pharmaceuticals

Note: The IPC-technology concordance table (available at: www.wipo.int/ipstats/en) was used to convert IPC symbols into 35 corresponding fields of technology.



31

Figure 4 reports inventor immigration rates for selected

technology fields for a number of countries.4 Generally,

countries such as Switzerland, the Netherlands and

the US had high inventor immigration rates in all of the

reported fields for the 2006-10 period. In contrast, China,

India and Japan reported low inventor immigration rates

for the same period. However, across countries and

technology fields, there were considerable variations in

inventor immigration rates.

Do regions play a role in attracting talent?

One striking aspect of immigration, and particularly

skilled immigration, is that migrants tend to concentrate

in specific geographical areas within countries. For

example, the share of skilled foreign-born individuals in

the UK and France in 2000 was estimated at 8.8% and

9.8%, respectively; in contrast, 28% of London residents

and 23% of Paris residents were foreign-born (Freeman,

2006). In particular, immigrant inventors appear to cluster

in metropolitan areas, thus contributing to the spatial

concentration of inventive activity. This issue is analysed

by matching PCT applications with the OECD’s REGPAT

database (Maraut et al, 2008; refer to Miguélez and

Raffo, 2013, for details of the matching procedure).5 By

linking inventor nationality information with REGPAT, it is

possible to study the settlement patterns of immigrant

inventors within countries beyond the settlement patterns

of native inventors.

4 The selection of technology fields was based on the

total number of PCT applications filed in 2010.

5 The latest version of REGPAT provides detailed

regional information on all EPO and PCT applicants,

and information on inventors for all OECD and EU

countries, as well as a few other selected countries.

Table 5 lists the top 20 European NUTS 2 regions in

terms of their inventor immigration rates.6 It shows that

European regions in highly innovative, middle-to-small

European countries ranked well above the European

average – although it should be noted that a few regions

of the UK - a large European country - appear in this list.

On the other hand, only six US states ranked above the

national average; these six were, however, regarded

as the most innovative and dynamic states. In order to

compare regions of similar size from Europe and the

US, it is worth repeating the analysis of the US data on

a more disaggregated level, such as in the Metropolitan

Statistical Areas (MSAs). In particular, some of the biggest

and most innovative MSAs – San Diego, San Jose-Santa

Clara, New York and Boston – appear in the top 20

ranking. When the MSA data are compared with the

European NUTS 2 data, one can see that the top four

European regions attract more talented individuals (in

relative terms) than does San Diego.

However, only few European NUTS2 regions had an

inventor immigration rate above 20%, while for the US

a larger number of MSAs reported immigration rates

greater than 20%. In other words, immigrant inventors’

settlement in European regions seemed to be more

skewed than was the case in the US.

6 NUTS stands for the French acronym “Nomenclature des unités territoriales statistiques”.


32

Table 5: Top 20 immigration rates by region, 2006-10

NUTS2 regionImmigration

rate (%) US statesImmigration

rate (%) US MSAsImmigration

rate (%)NORDWESTSCHWEIZ (CH) 50.7 CALIFORNIA 26.9 San Diego-Carlsbad-San Marcos, CA 36.7

RÉGION LÉMANIQUE (CH) 49.3 NEW JERSEY 24.2 Stockton, CA 33.3

RÉGION DE BRUXELLES (BE) 42.7 MASSACHUSETTS 21.8 Evansville, IN-KY 32.2

ZÜRICH (CH) 42.4 DELAWARE 21.2 Champaign-Urbana, IL 32.0

ZENTRALSCHWEIZ (CH) 36.0 NEW YORK 20.8 San Jose-Sunnyvale-Santa Clara, CA 31.0

LUXEMBOURG (LU) 35.7 TEXAS 18.9 Trenton-Ewing, NJ 30.4

OSTSCHWEIZ (CH) 31.0 MARYLAND 18.2 Albany-Schenectady-Troy, NY 28.5

PROV. BRABANT WALLON (BE) 30.1 CONNECTICUT 17.7 Columbus, IN 28.5

INNER LONDON (UK) 28.0 OREGON 17.4 Lansing-East Lansing, MI 28.3

SOUTHERN AND EASTERN (IE) 22.0 IDAHO 16.4 Athens-Clarke County, GA 28.2

PROV. LUXEMBOURG (BE) 21.5 HAWAII 16.1 Ithaca, NY 28.0

PROV. ANTWERPEN (BE) 19.7 FLORIDA 15.6 Ann Arbor, MI 27.7

OUTER LONDON (UK) 19.4 NEW MEXICO 15.4 Gainesville, FL 27.6

NOORD-BRABANT (NL) 19.3 ARKANSAS 15.1 College Station-Bryan, TX 27.3

ESPACE MITTELLAND (CH) 19.0 ILLINOIS 14.8 New York-Northern New Jersey-Long Island, NY-NJ-PA 24.3

PROV. VLAAMS-BRABANT (BE) 18.8 PENNSYLVANIA 14.6 Santa Barbara-Santa Maria-Goleta, CA 24.0

TICINO (CH) 18.2 GEORGIA 14.3 Ames, IA 23.2

TIROL (AT) 17.8 MICHIGAN 14.2 Dallas-Fort Worth-Arlington, TX 23.1

EAST ANGLIA (UK) 17.4 NORTH CAROLINA 14.1 State College, PA 22.6

PROV. HAINAUT (BE) 17.0 ARIZONA 13.9 Boston-Cambridge-Quincy, MA-NH 22.5

European average 9.7 US average 18.5 US average 18.5

Note: Only NUTS2 (Nomenclature des unités territoriales statistiques) regions with more than 25 native inventors and MSAs with more than 150 native inventors are listed here.


Table 6: Most populated migration corridors, 2006-10

Largest inventor migration corridors Largest inventor migration corridors (excluding the US)Origin Destination Inventors Origin Destination InventorsChina United States of America 27,698 Germany Switzerland 4,949

India United States of America 21,712 France Switzerland 1,879

Canada United States of America 11,363 France Germany 1,492

United Kingdom United States of America 8,314 China Japan 1,462

Germany United States of America 5,894 Germany Netherlands 1,332

Germany Switzerland 4,949 Austria Germany 1,307

Republic of Korea United States of America 4,876 France United Kingdom 1,210

France United States of America 3,901 China Singapore 1,149

Japan United States of America 2,843 Germany Austria 1,107

Russian Federation United States of America 2,308 United Kingdom Germany 1,080

France Switzerland 1,879 Netherlands Germany 1,049Israel United States of America 1,875 United States of America China 1,041

Australia United States of America 1,783 Germany United Kingdom 969

Netherlands United States of America 1,670 Italy Germany 956

Italy United States of America 1,492 Italy Switzerland 955

France Germany 1,492 France Belgium 934

China Japan 1,462 Germany France 916

Germany Netherlands 1,332 United Kingdom Switzerland 887

Austria Germany 1,307 United States of America Germany 820

Turkey United States of America 1,233 United States of America Canada 807



33

Which are the most popular inventor migration corridors?Table 6 shows the most populated bilateral corridors

during the 2006-10 period. The US emerged as the most

frequent destination country. Origin countries belong to

the high-income group, except China and India. The top

two corridors are China-US (27,698 inventors) and India-

US (21,712). In both cases, the high-income country is the

destination and the middle-income country is the origin.

When the US as a destination country was excluded from

the analysis, intra-European flows of inventors dominated

the top corridors. There were, however, some interesting

exceptions, such as the China-Japan (1,462) corridor and

the China-Singapore (1,149) corridor.

Asian countries – and to a lesser extent, countries from

Oceania – are important sources of inventors. Figure 5

depicts the top 10 most popular destinations for inven-

tors originating from the Middle East, South Asia, East

Asia and Oceania. As can be seen, the proportion of

inventors going to the US was greater than that going

to other countries. For example, close to nine times as

many migrant inventors from these regions as a whole

immigrated to the US (65,517) than immigrated to Europe

(7,660). They represented 55.9% of all immigrant inven-

tors in the US for the period 2006-10. While China’s and

India’s migration flows to the US were largely responsible

for this phenomenon, other countries also played a role.

Moreover, countries from the above-mentioned broad

geographical region featured among the top 10 destina-

tions for inventors. In particular, Australia, China, Japan,

Malaysia, New Zealand, Singapore and the Republic of

Korea attracted large numbers of inventors from this col-

lection of geographical regions. In addition, within Europe,

the UK received the largest share (28%) of inventors from

these regions, followed by Germany (24%).

For comparison purposes, Figure 6 depicts the top

10 most popular destinations for inventors from Latin

America and the Caribbean (LAC). As Table 1 shows,

the absolute number of inventors emigrating from the

LAC region was substantially lower when compared

with the corresponding figures for Asia. Again, the US

topped the ranking of destination countries. In relative

terms, LAC inventors accounted for approximately 3% of

all immigrants in the US and for approximately 2% of all

immigrants in Europe. Within Europe, Germany topped

the ranking (22% of all inventor migrants from the LAC re-

gion to Europe), and was followed by Switzerland, Spain

and France. A shared colonial heritage and a common

language explain why Spain attracted considerable talent

from LAC countries. The data also show considerable

intra-regional mobility of inventors within the LAC region.

For example, four LAC countries (Brazil, Chile, Colombia

and Mexico) are in the top 10 ranking as destination

countries for inventors originating from the LAC region.


34

Figure 5: Where do inventors from the Middle East, South Asia, East Asia, and Oceania emigrate from?


Figure 6: Where do LAC inventors emigrate from?



35

Do sending countries gain from the brain drain?

Despite the adverse consequences of the brain drain

of high-skilled people on a country’s potential develop-

ment, it is also well recognized that emigrants do not

necessarily sever their ties with their homelands and,

as diasporas, they may constitute a valuable resource

in terms of accessing foreign knowledge and technol-

ogies. To explore this point further, one can compute

the share of patents filed by the emigrant inventors of

each country that include at least one inventor residing

in the emigrant country of origin. The idea is to analyze

the extent to which each country’s emigrant inventor

community is committed to their country of origin and,

as a consequence, the extent of their collaboration with

their co-national colleagues at home. As the left axis of

Figure 7 shows, the US diaspora seems to be the most

committed to their homeland; 27.2% of the PCT applica-

tions with US inventor emigrants included US residents

among their co-inventors.

However, bearing in mind that the absolute number of

US resident inventors accounted for the world’s largest

number of resident inventors during the 2006-10 time-

frame, the probability of collaborating with a US resident

inventor was very high, regardless of the commitment of

US inventors abroad and the extent of their collaboration

with their home country colleagues. In order to illustrate

this last point, Figure 7 computes a hypothetical ratio

between the share of patents co-invented with nationals

of the country of origin and the share of total inventors

residing in the country of origin (see black and white

diamonds on right axis). The results show that inventors

from middle-income countries were actually the most

committed to their homelands, in that they collaborated

with their national colleagues at home disproportionately

more than would have been expected, given their share

of total inventors. In fact, the only two countries which

had a ratio lower than 1 during this period were China

and the US, which indicates that inventors from these

countries are less committed to their country of origin

than would have been expected, given their share of the

total number of inventors.

Figure 7: Share of PCT applications with homeland inventors and its ratio with the share of resident inventors with whom to collaborate: 2006-10

60

40

20

0

Ratio

hom

elan

d co

llabo

ratio

nsto

resid

ent i

nven

tors

0

10

20

30

Ratio

of P

CT a

pplic

atio

nsw

ith h

omel

and

inve

ntor

s (%

)

United

State

s of A

merica

German

yJap

anIsr

ael

Franc

e

Netherl

ands

United

King

dom

Russi

aIta

ly

Austral

ia

Canad

a

South

Africa

Repu

blic o

f Kore

aBra

zil

Mexico

China

India

Ukraine

Turkey

Malaysi

a

Origin

Applications' homeland inventors Ratio over countries' number of applications



36

Conclusion

This special section described a new global dataset on

migrant inventors, using information on inventor nation-

ality and residence gleaned from PCT applications. From

this analysis, two important facts emerged. First, from a

methodological perspective, this section demonstrated

that PCT data are meaningful and are useful in analyzing

the interplay between migration and innovation. Second,

from a more analytical viewpoint, the data reveal a

number of interesting findings that are worth highlighting.

From the methodological perspective, use of patent data

to map the migratory patterns of high-skilled workers can

address some of the limitations associated with existing

migration datasets. In particular, this database covers

a long time period, provides information on an annual

basis, and contains data for a large number of sending

and receiving countries. Inventors constitute a group of

high-skilled workers of special economic importance

who have more homogenous skills than tertiary-educated

workers as a whole.

Broadly speaking, the data clearly demonstrate that

the pattern of inventors’ mobility resembles other high-

skilled migration figures, and in particular, what we know

about the migration of scientists and engineers based

on anecdotal evidence, surveys and media reports.

For example, the majority of immigrant inventors in the

2006-10 period were concentrated in the US, whereas

European countries lagged behind in this respect. The

US not only had the largest absolute number of immi-

grant inventors during this period, but it also stood out

as one of the main receiving countries relative to its total

population of inventors.

The data highlight important differences across countries

as well as within countries and across different cities,

technologies and organizations employing inventors (ap-

plicants). In addition, they highlight that during the 2006-

10 period, immigration rates were remarkably different

across applicant types i.e., university, government and

research institutions, business, and individuals. Within

these groupings, university/government immigration

rates were considerably higher than business sector

immigration rates. In relation to data for fields of technol-

ogy, for example, during the 2006-10 period immigration

rates varied from 4.1% (mechanical elements) to 18.3%

(micro-structure and nano-technology). Other fields

also relied heavily on immigrant inventors; such fields

included pharmaceuticals (14.6%), biotechnology (14.6%),

digital communication (15.2%) and basic communication

processes (16%).

Furthermore, by using unit record data, it becomes

possible to link patent-inventor data with citation and

co-inventorship information. It also becomes possible

to study social relationships between inventors and sub-

sequent knowledge diffusion patterns across countries,

regions and technology fields. Additionally, data can also

be linked to country-, city- and firm-level information in

order to provide new empirical evidence on a broad

range of interrelated topics.

From an analytical standpoint, this special section pro-

vides new evidence on the migration patterns of knowl-

edge workers which, to date, have probably not received

the attention that this subject deserves. As a result, most

analysis on the migration patterns of scientists and engi-

neers has exclusively focused on the US experience and

its major providers of foreign talent, namely China and

India (Breschi et al, 2013). However, high-skilled worker

migration is a multipolar phenomenon, implying a large

number of sending and receiving countries.


37

Thus, for example, it is possible to observe trends in

important talent circulation between Western European

countries during the 2006-10 period. It is also possible

to observe that the number of non-European countries

providing talent to Europe did not necessarily coincide

with migration flows to the US – e.g., from African or

LAC countries. During this period, European countries

also constituted the main providers of talent to the US.

There is large “brain circulation” between Asian econo-

mies, with Singapore standing out as a major receiving

country. For its part, China is a major provider of talent

within its geographical area of influence; however, in

recent years, it has also attracted a large number of

immigrant inventors, both from Asia and the rest of the

world. Finally, albeit to a lesser extent, migrant inventors

also originate in other areas of the world, such as LAC

countries and Africa.

Of course, using patent data for the purpose of economic

analysis does not come without limitations. One import-

ant caveat is that one only observes inventors when they

seek patent protection. Not all inventions are patented,

however, and there is no one-to-one correspondence

between the number of patent applications filed and the

commercial value of the underlying inventions or their

contribution to technological progress. Another limitation

is that the PCT dataset does not include inventors with a

migratory background who have become a host country

national. Unfortunately, the data do not facilitate the as-

sessment of how severe these biases are. In using these

data, one should be aware of such limitations, especially

when drawing policy conclusions.

Notwithstanding these caveats, this new database mean-

ingfully captures a phenomenon of growing importance.

Indeed, the descriptive overview presented in this section

suggests that it is consistent with migratory patterns

and trends elicited from census data. At the same time,

the database opens up new avenues for research and

promises to generate fresh empirical insights that can

inform both innovation policy and migration policy.


38

References

Black, G.C., Stephan, P.E., 2008. “The Economics of

University Lab Science and the Role of Foreign Graduate

Students and Postdoctoral Scholars” (SSRN Scholarly

Paper No. ID 1323429). Social Science Research

Network, Rochester, NY.

Breschi, S., Lissoni, F., Tarasconi, G., 2013. “Inventor

Data for Research on Migration and Innovation: A Survey

and a Pilot. Forthcoming as a WIPO Economic Working

Research Paper.

Fink, C., Miguélez, E., Raffo, J., 2013. The global race

for inventors. www.voxeu.org/article/global-race-inven-

tors

Freeman, R.B., 2006. “People Flows in Globalization”.

The Journal of Economic Perspectives 20, 145–170.

Kerr, W., 2009. “The Agglomeration of US Ethnic

Inventors” (Working Paper No. 15501). National Bureau

of Economic Research.

Kerr, W.R., 2013. “U.S. High-Skilled Immigration,

Innovation, and Entrepreneurship: Empirical Approaches

and Evidence” (Working Paper No. 19377). National

Bureau of Economic Research.

Maraut, S., Dernis, H., Webb, C., Spiezia, V., Guellec,

D., 2008. “The OECD REGPAT Database” (OECD Science,

Technology and Industry Working Papers). Organisation

for Economic Co-operation and Development, Paris.

Miguélez, E., Fink, C., 2013. “Measuring the

International Mobility of Inventors: A New Database”

(WIPO Economic Research Working Paper No. 8). World

Intellectual Property Organization – Economics and

Statistics Division.

Miguélez, E., Raffo, J., 2013. “The spatial distribution

of migrant inventors”. Forthcoming as a WIPO Economic

Research Working Paper.

WIPO, 2012. World Intellectual Property Indicators, 2012

edition (WIPO Economics and Statistics Series). World

Intellectual Property Organization – Economics and

Statistics Division.

special section the international mobility of inventors … · special section the international...

Documents